yujunjun/olmocr
1
0
Fork 0
mirror of https://github.com/allenai/olmocr.git synced 2025-09-22 06:57:59 +00:00
Code Issues Packages Projects Releases Wiki Activity

Oops removing print

Browse Source
This commit is contained in:
Jake Poznanski 2024-10-11 16:23:14 +00:00
parent aea3f7f1fe
commit 2dccc4be3b
1 changed files with 0 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
pdelfin/train/dataprep.py
View File

@ -60,8 +60,6 @@ def prepare_data_for_qwen2_training(example, processor, add_batch_dim=False):
return_tensors="np" return_tensors="np"
) )
print(labels["input_ids"].shape)
# Append an <|im_end|>\n" to the labels, because this is what it would look like # Append an <|im_end|>\n" to the labels, because this is what it would look like
# if we passed the whole message stream in there # if we passed the whole message stream in there
im_end_tokens = processor.tokenizer("<|im_end|>\n", add_special_tokens=False)["input_ids"] im_end_tokens = processor.tokenizer("<|im_end|>\n", add_special_tokens=False)["input_ids"]